The goal of this application is to exploit the experimental merits of genetically engineered mouse models (GEMMs) of cancer for use in basic and translational cancer science. This application will make use of high refined mouse models of malignant melanoma and prostate cancer harboring signature mutations that are present in the human disease. These models recapitulate the biological, pathological and clinical features of the human disease including metastasis. In Program 1, inducible NRAS* metastatic melanoma model and a high metastatic PTEN/SMAD4 prostate cancer model will be engineered to experience human-like cancer genome instability via incorporation of telomerase deficiency. Using a full range multi-dimensional genome profiling methodologies, we will examine the acquired epigenetic and genetic events in primary and metastatic tumors as well as tumors that have escaped dependence on NRAS* signaling in melanoma or androgen-dependence in prostate cancer. By comparison with human cancer genome atlas profiles, these comparative oncogenomics studies will serve to facilitate the discovery of key driver events in human cancer and understand the nature of these mutational events in the context of metastasis, RAS signaling and androgen dependence. In Program 2, we will utilize our mouse models of metastatic melanoma and prostate cancer to identify metastatic determinants that can predict which early stage tumors are programmed to remain indolent versus pursue an aggressive metastatic course. Specifically, we will take advantage of our GEMMs with differing metastatic potential for each tumor type to develop a metastasis mRNA signature. These murine data will then be interfaced with human DNA copy number data to identify cross-species conserved metastasis-associated genomic events that can then be enlisted into functional validation screens to identify biologically active metastasis determinants. These functional metastasis determinants will be tested in outcome-annotated clinical samples for their predictive value and capacity to guide the optimal clinical management of early stage melanoma and prostate cancers.